Synthesis and evaluation of a set of 4-Phenylpiperidines and 4-Phenylpiperazines as D2 receptor ligands and the discovery of the dopaminergic stabilizer 4-[3-(Methylsulfonyl)phenyl]-l-propylpiperidine (huntexil, pridopidine, ACR16)

Article abstract of DOI:10.1021/jm901689v

Modification of the partial dopamine type 2 receptor (D₂) agonist 3-(l-benzylpiperidin-4-yl)phenol (9a) generated a series of novel functional D₂ antagonists with fast-off kinetic properties. A representative of this series, pridopid

Full text of DOI:10.1021/jm901689v

2510 J. Med. Chem. 2010, 53, 2510–2520
DOI: 10.1021/jm901689v
Synthesis and Evaluation of a Set of 4-Phenylpiperidines and 4-Phenylpiperazines as D2
Receptor Ligands and the Discovery of the Dopaminergic Stabilizer
4-[3-(Methylsulfonyl)phenyl]-1-propylpiperidine (Huntexil, Pridopidine, ACR16)
ꢀ
Fredrik Pettersson,* Henrik Ponten, Nicholas Waters, Susanna Waters, and Clas Sonesson
NeuroSearch Sweden AB, Arvid Wallgrens Backe 20, S-413 46 Go€teborg, Sweden
Received November 16, 2009
Modification of the partial dopamine type 2 receptor (D2) agonist 3-(1-benzylpiperidin-4-yl)phenol (9a)
generated a series of novel functional D2 antagonists with fast-off kinetic properties. A representative of
this series, pridopidine (4-[3-(methylsulfonyl)phenyl]-1-propylpiperidine; ACR16, 12b), bound compe-
titively with low affinity to D2 in vitro, without displaying properties essential for interaction with D2 in
the inactive state, thereby allowing receptors to rapidly regain responsiveness. In vivo, neurochemical
effects of 12b were similar to those of D2 antagonists, and in a model of locomotor hyperactivity, 12b
dose-dependently reduced activity. In contrast to classic D2 antagonists, 12b increased spontaneous
locomotor activity in partly habituated animals. The “agonist-like” kinetic profile of 12b, combined
with its lack of intrinsic activity, induces a functional state-dependent D2 antagonism that can vary with
local, real-time dopamine concentration fluctuations around distinct receptor populations. These
properties may contribute to its unique “dopaminergic stabilizer” characteristics, differentiating 12b
from D2 antagonists and partial D2 agonists.
Introduction
Activation and G-protein coupling transform the D2 re-
ceptor to a state in which dopamine binds with higher affinity.
Thus, the D2 receptor population is distributed between (i) a
resting, low-affinity state (D2Low) and (ii) a catalytically
active, high-affinity state (D2High).4,5 The D2 agonists dopa-
mine (1, Figure 1) and apomorphine (2, Figure 1) display high
affinity in agonist ligand binding assays and induce a full
catalytic reaction in functional assays (i.e., they have affinity
and high intrinsic activity).6-8 Dopamine D2 receptor partial
agonists also generally bind with high affinity in such assays
but are associated with less intrinsic activity than full agonists.
Finally, dopamine D2 receptor antagonists, such as haloper-
idol (3, Figure 1) and olanzapine (4, Figure 1), bind with equal
affinity in agonist and antagonist ligand binding assays4,5,7
while essentially lacking intrinsic activity. The ratio of antago-
nisttoagonist binding affinity [Ki(D2Low)/Ki(D2High)] hasthus
been used as a measure of intrinsic activity.6,9-11 It has been
speculated that the main part of auto- and presynaptic
Dopamine type 2 receptors (D2) are primarily located in the
basal ganglia of the mammalian brain but also occur in other
structures of the brain, such as the cortex. The receptors,
which are located at the neuronal membrane, belong to
the monoamine subclass of the G-protein-coupled seven-
transmembrane receptors (GPCRs).1 In the brain, dopamine
(1, Figure 1) exerts its action by means of synaptic as well as
extrasynaptic release, affecting postsynaptic, presynaptic, and
dendritic D2 receptor populations. Synaptic dopamine release
is followed by fast reabsorption or degradation, processes that
terminate D2 signaling. Drugs that interact with the agonist
binding site of D2 receptors can be described as antagonists,
partial agonists, or full agonists, and a number of these drugs
have well-established applications in the treatment of various
neurological and psychiatric disorders.
The affinity ofa drugbinding toits receptor isdependent on
2
its association and dissociation rate constants, kon and koff
.
receptors, which control the synthesis and release of dop-
High
Commonly, medicinal chemistry optimization programs have
generated high-affinity drugs with slow drug-receptor ki-
netics. Limited attention has been devoted toward optimizing
D2 ligands with receptor kinetics comparable to those of
natural dopamine signaling. It has been shown that dopamine
D2 receptor kinetics differ among antipsychotic compounds,
and it is proposed that fast-off kinetics (high koff) is associated
with less extrapyramidal side effect (EPS) liability,3 probably
because physiological responses to normal dopamine surges
are possible.
amine, are of the dopamine D2
synaptic receptors, on the other hand, are more equally
affinity state. The post-
High
Low
and D2
distributed between the D2
affinity states
depending on the concentration of dopamine in the synapse.
Further, it has been proposed that auto- and presynaptic
receptors are sensitive to lower levels of endogenous dop-
amine than postsynaptic receptors. This difference in sensitivity
to endogenous dopamine levels might explain why D2 partial
agonists behave primarily as agonists at auto- and presynaptic
receptors and as antagonists at postsynaptic receptors.12
Traditional dopamine D2 antagonists, such as haloperi-
dol and olanzapine, are used to treat positive symptoms
in schizophrenia and manic episodes in bipolar disorder.
*To whom correspondence should be addressed. Phone: +(46)
31 7727710. Fax: +(46) 31 7727701. E-mail: fredrik.pettersson@
neurosearch.se.
r
pubs.acs.org/jmc
Published on Web 02/15/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2511
Figure 1. Dopamine D2 ligands. Dopamine D2 receptor agonists dopamine (1) and apomorphine (2), classical antagonists haloperidol (3) and
olanzapine (4), partial agonists (-)-3-(3-hydroxyphenyl)-N-n-propylpiperidine (5), bifeprunox (6), aripiprazole (7), and 3-(1-benzylpiperidin-
4-yl)phenol (9a), and dopaminergic stabilizers S-(-)-OSU6162 (8) and pridopidine (12b).
These compounds are associated with EPS, believed to occur
as a result of excessive attenuation of brain dopamine neuronal
activity due to the blockage of postsynaptic D2 receptors.13
The partial dopamine D2 agonists, (-)-3-(3-hydroxyphenyl)-
N-n-propylpiperidine ((-)-3PPP; 5, Figure 1), bifeprunox
(6, Figure 1) and aripiprazole (7, Figure 1) are effective in
treating both the positive and negative symptoms of schizo-
phrenia and have been reported to have a lower EPS liability
than both typical and atypical antipsychotics.13-15 Because of
the low intrinsic activity of aripiprazole, it is thought to act as
either a functional agonist or a functional antagonist, depend-
ing on the initial levels of dopamine; aripiprazole has been
labeled a “dopamine system stabilizer”.16
Froma medicinal chemistry perspective, generally when the
chemical properties of the agonists for a specific receptor
system are compared with those of the corresponding antago-
nists, the agonists are relatively small molecules and hydro-
philic in character, whereas the corresponding antagonists are
usually larger and more lipophilic, lacking the essential phar-
macophore elements for displaying agonist properties.17-24
This general principle is illustrated by the two dopamine D2
agonists dopamine and apomorphine (1 and 2, Figure 1),
which have calculated logP (clogP)25 values of 0.17 and
2.49, respectively, compared with the corresponding anta-
gonists haloperidol and olanzapine (a structural analogue
of clozapine) (3 and 4, Figure 1), which have clog P values of
4.45 and 4.51, respectively. The antagonists also lack certain
essential pharmacophore elements, such as the catechol
group and the basic nitrogen with the correct conformation
and distance (i.e., phenethylamine) from the aromatic moiety.
The sites of interaction between agonists and the dopamine
D2 receptor have been characterized by mutagenesis studies in
combination with three-dimensional homology modeling.
This has revealed that Asp-118 on the third TMa helix is
important for formation of a salt bridge with the protonated
nitrogen, that serine residues in TM5 (Ser-193, Ser-194, and
Ser-197) are important for formation of hydrogen bond
interactions with the catechol function, and that a cluster of
aromatic residues in TM4 and TM6 contribute tostabilization
of the drug-receptor complex via hydrophobic interactions
(mainly π-π stacking).26-33 It has been proposed that TM6
undergoes a translational or rotational movement in the
activation phase of GPCRs and that the interaction with an
agonist facilitates this movement.34-37 In line with this pro-
posal, Goddard et al.38 have speculated that interactions with
TM3 (Asp-118) and TM5 (Ser-193 and Ser-197) by dopamine
D2 agonists pulls TM3 and TM5 closer together in the
active state, allowing the flexible motion of TM6. By con-
trast, an antagonist (such as haloperidol) interacts strongly
with TM helices 3 and 6 (having minimal contact with TM5),
which therefore prevents such movement.33,38 Interestingly,
Goddard et al. have suggested that, in contrast to haloperidol,
aAbbreviations: EPS, extrapyramidal side effects; clog, calculated log;
TM, transmembrane; Asp, aspartic acid; Ser, serine; SAR, structure-
activity relationship; m-CPBA, m-chloroperoxybenzoic acid; Emax, max-
imal efficacy; LMA, locomotor activity; HPLC/EC, high-performance
liquid chromatography analysis with electrochemical detection; DOPAC,
3,4-dihydroxyphenylacetic acid; HVA, homovanillicacid; SEM, standard
error of the mean; logD, the apparent log P values for pH 0-14; LClogD,
liquid chromatography-based logD ; TFA, trifluoroacetic acid.

2512 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Pettersson et al.
clozapine binds to the agonist-binding pocket but lacks the
tight interaction with serine residues on TM5 that is essen-
tial for the conformational changes necessary for intrinsic
activity.36 Thus, the more lipophilic antagonists seem to
interact with hydrophobic residues in the binding pocket,
which stabilizes the inactive state and prevents the optimal
conformational changes as described above for agonists.
Therefore, a useful approach in developing antagonists from
agonist or partial agonist structures is to add a lipophilic
system, such as a phenyl or cyclohexyl group, in a specific
position on the agonist/partial agonist structure. This group
could then interact with a hydrophobic part of the receptor
system, preventing the receptor from undergoing the con-
formational changes needed for agonism.22,23,39-42
In the search for novel dopamine D2 receptor antagonists,
we have used a different approach, focusing on identifying
the critical elements in D2 agonists/partial agonists that
are essential for intrinsic activity, under the hypothesis that
a careful modification of the physicochemical properties of
these elements would generate compounds with antagonistic
properties (i.e., no intrinsic activity). The key to this approach
was to maintain the chemical backbone of the agonist/partial
agonist and performing the modification in such a way that
the hydrophilicity either is retained or is even higher after the
modification. We speculated that such modifications might
lead to compounds that antagonize the actions of dop-
amine but, unlike lipophilic antagonists, lack the ability to
stabilize the inactive state (D2Low) of the D2 receptor. Further,
we speculated that these compounds could exert modula-
tory effects on dopamine transmission and possibly state-
dependent activity in vivo.
We have previously reported studies on a series of ana-
logues to the 3-substituted phenylpiperidine 5, one of the first
reported partial D2 agonists,12,43 in which the validity of this
approach was investigated. One of the key elements for the
intrinsic activity of 5 has been shown to be a hydrogen bond
donating/accepting phenolic OH group.44 In a structure-
activity relationship (SAR) study of these 3-substituted
phenylpiperidines, we discovered that replacing the 3-OH
group in 5 with electron-withdrawing groups rendered com-
pounds with antagonistic properties at the dopamine D2
receptor.44,45 One of these compounds, 8 (Figure 1), has been
found to display unique effects in vivo. Neurochemically, 8
displays effects similar to those of classic dopamine D2
antagonists, such as an increase in synthesis and turnover of
dopamine. However, in sharp contrast to the D2 antagonists,
8 can stimulate, suppress, or show no effect on motor and
behavioral symptoms, depending upon the prevailing dop-
aminergic tone. Therefore, the effects on motor and beha-
vioral symptoms have been regarded as state dependent and 8
has been classified as a “dopaminergic stabilizer”.45-47 Thus,
8 represents a dopamine D2 antagonist developed from a
partial agonist, without the addition of lipophilic ring systems.
It should be mentioned that 5 has a clog P of 3.18 whereas 8
has a clogP of 2.21. From in vitro binding studies, it has been
shown that 8 more potently displaces an “agonist” from the
D2High binding site than an antagonist from the D2Low binding
site [a Ki(D2Low)/Ki(D2High) ratio of 14 and 137 has been
reported],45,48 which would suggest that 8 has some intrinsic
activity, and it has indeed been reported that 8 can have minor
agonist effect at D2 receptors in specially designed in vitro
systems.48,49 However, it has been clearly shown that 8 does
not display any intrinsic activity in vivo, and therefore, the
predictive value of these in vitro findings of intrinsic activity
Table 1. Compounds 9a-12b
compd
X
R
R0
9a
9b
N
OH
OH
benzyl
benzyl
propyl
propyl
benzyl
benzyl
propyl
propyl
CH
N
10a
10b
11a
11b
12a
12b
OH
OH
CH
N
SO2Me
SO2Me
SO2Me
SO2Me
CH
N
CH
requires further consideration.45,47,50 The preference of dopa-
minergic stabilizers, such as8, for the displacement of agonists
rather than antagonists at dopamine D2 receptors triggered
us to investigate the ability of various ligands to inhibit the
D2 receptor at different dopamine concentrations. We have
recently demonstrated that haloperidol and aripiprazole dis-
play insurmountable (noncompetitive) D2 antagonism where-
as compounds such as 8 display surmountable (competitive)
D2 antagonism.50 The insurmountable antagonism mediated
by haloperidol and aripiprazole strongly indicates that they
induce a long-lasting interaction with dopamine D2 receptors,
preventing dopamine from binding to and activating the
receptor. The dopaminergic stabilizer, 8, displays surmoun-
table antagonism of dopamine, since it lacks the physico-
chemical properties required to stabilize the inactive state.
Dopaminergic stabilizers have also been shown to display
fast dissociation from D2 receptors, which is believed to allow
the dopamine receptors to rapidly regain responsiveness
to dopamine.50 These differences regarding mode of action
may result in the unique effects of dopaminergic stabilizers
in vivo, which are not seen with D2 antagonists, agonists, or
partial agonists.
In the search for new chemical scaffolds to serve as starting
points for the development of dopaminergic stabilizers, we
have for the present study focused our attention on a series of
dopamine D2 partial agonists thatwerereported by Mewshaw
and colleagues51,52 including 3-(1-benzylpiperidin-4-yl)phe-
nol (9a, Figure 1, Table 1). This compound has been shown
to bind with high preference for the activated state of the D2
receptor (D2High) and has been classified as a potential partial
agonist based on the Ki(D2Low)/Ki(D2High) ratio. We specu-
lated that the key elements of 9a responsible for its intrinsic
activity were the phenol group, the anilinic nitrogen, and the
large N-alkyl group and that modifications of these groups
might lead to compounds with dopaminergic stabilizer prop-
erties. We herein present and discuss the synthesis and SAR,
based on in vitro and in vivo data, of a series of novel
compounds designed according to these principles. We also
describe the receptor dissociation kinetics at the dopamine
D2 receptor of both these compounds and a set of known
D2 ligands.
Chemistry
Compound 13 was synthesized from 1-bromo-3-methyl-
thiobenzene by lithiation with n-butyllithium and quench-
ing with N-substituted 4-piperidone (Scheme 1). Subsequent
treatment with trifluoroacetic acid (TFA) in a CH2Cl2

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2513
Scheme 1. Synthesis of Piperidines 11b and 12ba
a Reagents and conditions: (a) n-butyllithium, 1-Boc-4-piperidone, THF; (b) trifluoroacetic acid, CH2Cl2, Δ; (c) triethylamine, methyl chloroformate,
CH2Cl2; (d) m-CPBA, CH2Cl2; (e) Pd/C, H2, MeOH, HCl; (f) HCl, EtOH, Δ; (g) RX, K2CO3, acetonitrile, Δ.
Scheme 2. Synthesis of Phenols 9a-10ba
Scheme 3. Synthesis of Piperazines 11a and 12aa
a Reagents and conditions: (i) RX, K2CO3, acetonitrile, Δ.
solution gave 14 in excellent yield. Since sulfides contaminate
the Pd in the catalyst used to reduce the piperidene double
bond,53 14 was first protected by the addition of methyl
chloroformate to afford the carbamate 15 and then quanti-
tatively oxidized to the corresponding sulfone 16 using
m-chloroperbenzoic acid (m-CPBA). The protection of the amine
functionality was essential to avoid oxidation of the tetrahy-
dropyridine to pyridine. The sulfone 16 was reduced with
catalytic hydrogenation (Pd/C), affording the piperidine deri-
vative 17 in good yield. After deprotection of 17 with aqueous
HCl (8 M), the secondary amine 18 was treated with iodopro-
pane and benzyl bromide, respectively, affording the desired
products 11b and 12b in moderate to good yields (41-79%)
(Scheme 1). Commercially available 3-(piperidin-4-yl)phenol
was alkylated in the same manner, affording 3-(N-propyl-
piperidin-4-yl)phenol (10b) and 3-(N-benzylpiperidin-4-yl)-
phenol (9b, Scheme 2). Compound 19 (1-[3-(methylsulfonyl)-
phenyl]piperazine) was afforded in moderate yield by C-N
palladium-catalyzed cross-coupling of 1-bromo-3-(methylsul-
fonyl)benzene with piperazine using Pd2(dba)3 and rac-BI-
NAP in toluene (Scheme 3). Reacting 19 or commercially
available 1-(3-hydroxyphenyl)piperazine with iodopropane
and benzyl bromide, respectively, afforded the alkylated pipe-
razines 9a, 10a, 11a, and 12a in an average yield of 50%
(Schemes 2 and 3).
a Reagents and conditions: (i) piperazine, NaOtBu, Pd2(dba)3, rac-
BINAP, toluene, Δ; (ii) RX, K2CO3, acetonitrile, Δ.
Results
In Vitro. The agonists dopamine and apomorphine (1 and
2, respectively, Figure 1) bound with a strong preference to
the agonist site (D2High) and possessed full intrinsic activity
(Table 2). The classical D2 antagonist, haloperidol (3), did
High
Low
not discriminate between the D2
whereas olanzapine (4) seemed to have a slightly higher
and D2
states,
High
affinity for D2
than for D2Low. The partial agonist
High
(-)-3-PPP (5) bound with 130-fold higher affinity to D2
Low
than to D2
and displayed some, but not full, intrinsic
activity (26% maximal efficacy [Emax]). Bifeprunox (6) had a
binding ratio of only 2.5 but displayed an intrinsic activity
(Emax= 27%) similar to that of (-)-3-PPP in the functional
assay. Aripiprazole (7) has previously been reported either to
have a higher affinity for the agonist site than for the antagonist
site54 or to have similar affinity for both sites.55 Our data sup-
port the findings of Tadori et al.55 who reported that aripipra-
zole had no preference for D2High, but we found that the drug
also displayed 11% intrinsic activity (Table 2). Results from our
in vitro screening confirmed the binding profile previously

2514 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Pettersson et al.
Table 2. LClogD, clog P, and in Vitro Data for Compounds 9a-12b and Reference Compounds
LClogDa
clog P
D2Low Ki (nM)b
D2High Ki (nM)b
KiLow/ KiHigh
Emax
c
compd
9a
9b
4.93
3.93
4.22
1.81
5.32
4.39
3.19
1.86
NT
3.21
3.91
2.55
3.18
2.67
2.94
2.00
2.21
0.17
2.49
3.85
3.01
3.33
4.78
5.31
2.36
115 (76-173)
4.6 (3.7-5.7)
35 (28-43)
69 (49-97)
25
36 ( 8
0
285 (119-678)
721 (424-1225)
2570 (1270-5181)
1923 (899-4113)
841 (522-1355)
1211 (766-1915)
17550 (4588 - 67150)
2100 (1126-3919)
244 (127-467)
1.9 (1.7-2.2)
8.1
10
10a
10b
11a
11b
12a
12b
0
349 (278-438)
431 (297-626)
392 (265-580)
664 (498-886)
7521 (4057-13940)
2.9 (1.71-4.90)
0.62 (0.49-0.79)
1 (0.7-1.4)
6.1 (4.4-8.4)
17.4 (13-22)
0.04 (0.03-0.06)
2.8 (2-4.1)
7.3
4.4
2.1
1.8
2.3
724
393
1.9
5.2
130
2.5
0.9
5.1
0
0
0
0
0d
100d
91 ( 6d
0d
dopamine (1)
apomorphine (2)
haloperidol (3)
olanzapine (4)
(-)-3PPP (5)
bifeprunox (6)
aripiprazole (7)
8
3.43
4.45
4.51
1.91
6.59
6.55
2.22
32 (17-62)
NT
2268 (1203-4278)
0.1 (0.06-0.17)
2.6 (2-3.2)
23 ( 8d
28 ( 3d
11 ( 2d
0d
3884 (2304-6545)
755 (489-1166)
a Calculated from liquid chromatography retention time. b Binding affinities (apparent Ki) of selected compounds to recombinant HEK-293 cells with
[3H]spiperone as ligand for D2Low and [3H]7-OH-DPAT for D2High (95% confidence interval in parentheses). c Percentage of maximal efficacy (Emax
values on D2L-GRqi5 HEK293 cells,50 n = 3-4. NT: not tested. d Data from Dyhring et al.50
)
Figure 2. Recovery of dopamine D2 receptor-mediated responsive-
ness after ligand washout. HEK-hD2L-GRqi5 cells were pretreated
for 5 min with test compound, washed, and incubated for 5-
120 min at room temperature. In order to ensure maximal receptor
occupancies, high test compound concentrations (approximately
30-fold higher than the EC50/IC50 value) were used. After the
recovery period, fluorescence responses were measured in response
to 300 nM dopamine and normalized to the dopamine response in
the absence of compound pretreatment (presented as a gray line at
Figure 3. Dose-dependent effects of 12b and several dopamine D2
agonists, antagonists, and partial agonists on DOPAC levels in the
striatum.
100%). Data points represent the mean ( SEM, n=3-4.
reported by Mewshaw et al. for compound 9a (Table 2;
Ki(D2High)=4.6 nM, Ki(D2Low)=115 nM).52 In addition, 9a
displayed partial agonist activity in the functional assay, with an
efficacy of 36%. Chemical modifications of 9a produced a
group of compounds (9b-12b) with lower affinity for both
striatum (Figure 3 and Table 3) and a pronounced increase in
LMA in partly habituated rats, whereas the classic dopamine
D2 antagonists, haloperidol and olanzapine, induced sub-
stantial increases in DOPAC levels (∼360-420% of control
levels) along with strong (haloperidol) to moderate
(olanzapine) reductions in LMA (Figure 3 and Table 3).
The partial agonists (-)-3-PPP, aripiprazole, and bifepru-
nox induced a dose-dependent, partial increase in DOPAC
levels relative to haloperidol/olanazapine (∼150-175% of
control levels, Figure 3). In addition, aripiprazole and bife-
prunox induced a strong reduction in spontaneous LMA in
partly habituated rats, whereas (-)-3-PPP induced only a
mild reduction (Table 3). The antagonists and partial ago-
nists were also effective in decreasing amphetamine-induced
hyperlocomotion (Table 3). In contrast to the agonists,
partial agonists, and antagonists described above, the dop-
aminergic stabilizer 8 generated an increase in DOPAC
levels comparable to that seen with classic antagonists but,
High
D2
Low
and D2
but with a consistent preference for the
agonist site (D2High, Table 2). However, none of these com-
pounds displayed any detectable intrinsic activity in the func-
tional in vitro assay (Table 2), in the case of 12b consistent with
previous reports.47,56 Regarding receptor kinetics, compounds
9a-12b all displayed fast dissociation from the dopamine D2
receptor as shown in Figure 2 and as reported by Dyhring et al.50
The D2 recovery rates measured for these compounds were
similar to that of dopamine and in sharp contrast to those of
classic antagonists, such as haloperidol, and partial agonists,
such as aripiprazole and bifeprunox. The atypical anti-
psychotic olanzapine on the other hand has been shown to
display a moderate dissociation rate.50
In Vivo. The dopamine agonist, apomorphine, induced a
dose-dependent reduction in tissue levels of DOPAC in the

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2515
Table 3. In Vivo Data for Compounds 9a-12b and Reference Com-
pounds
The binding properties of the lead compound, 9a, reported
by Mewshaw et al. to be a partial agonist,52 were confirmed,
and an intrinsic activity of 36% further supports its classifica-
tion as a partial agonist. In addition, 9a was found to display
surmountable D2 antagonism against dopamine (competitive
effect) and fast receptor dissociation kinetics in vitro. All
structural modifications of 9a made in the present study
(9b-12b; Tables 1 and 2) resulted in compounds with no
intrinsic activity but with a consistent binding preference for
D2High, albeit less pronounced than that of 9a, indicating that
all the structural elements in 9a (phenol, piperazine, and
N-benzyl) are required for intrinsic activity at the D2 receptor.
These findings are in contrast to the 3-substituted phenyl-
piperidines (e.g., 5 (3-PPP)), which can accept a larger varia-
tion in substitution pattern on the aromatic ring and the basic
nitrogen while maintaining high intrinsic activity.45,57 This
indicates that the lack of phenethylamine backbone in the 9a
series leads to a structural class less prone to induce intrinsic
activity, and therefore, additional interactions are needed.
In addition, compounds 9b-12b were found to display
surmountable D2 antagonism with fast dissociation from
the dopamine D2 receptor (Table 2 and Figure 2), findings
in line with what we would expect from these compounds as a
result of the similarity to the agonist chemical motif and their
high hydrophilicity.
DOPAC, %
of control (
SEMa
LMA, % of
control (
SEMb
LMA, % of
amphetamine
( SEMc
compd
9a
9b
108 ( 4d
8 ( 2d
10 ( 2d,m
12 ( 1e,m
16 ( 1e,m
15 ( 4d,m
6 ( 2f,m
6 ( 1e,m
12 ( 3g,m
NT
1 ( 0i,m
6 ( 2j,m
NT
5 ( 3l,m
9 ( 1l,m
37 ( 14d
317 ( 13e,m
261 ( 39e,m
248 ( 10d,m
310 ( 16e,m
310 ( 16e,m
265 ( 10d,m
77 ( 3h,m
85 ( 25e
31 ( 3e
10a
11a
11b
12a
12b
17 ( 9d
44 ( 3e
27 ( 9e
262 ( 28d,m
7118 ( 779h,m
7 ( 4i
apomorphine (2)
haloperidol (3)
olanzapine (4)
(-)-3PPP (5)
bifeprunox (6)
aripiprazole (7)
8
425 ( 3i,m
362 ( 5j,m
175 ( 8k,m
152 ( 12l,m
150 ( 3l,m
260 ( 15d,m
17 ( 5j,m
37 ( 17k
7 ( 2l,m
2 ( 1l,m
215 ( 62d
a Post-mortem biochemistry of levels of DOPAC in the striatum
compared to saline control (n = 4). b LMA during 15-60 min after
injection measured at 25 Hz compared to saline control. c LMA during
15-60 min after injection measured at 25 Hz compared to amphetamine
(1.5 mg/kg intraperitoneally) pretreated rats (n = 4). To compare the in
vivo effects of the different compounds, the lowest dose required
to produce a maximal DOPAC response was selected. NT: not tested.
d 100 μmol/kg. e 33 μmol/kg. f 50 μmol/kg. g 150 μmol/kg. h 6.59 μmol/kg
(2 mg/kg). i 1.0 μmol/kg (0.37 mg/kg). j 10.5 μmol/kg (3.3 mg/kg).
k117 μmol/kg (30 mg/kg). l 7.68 μmol/kg (3.7 mg/kg). l 4.46 μmol/kg
(2 mg/kg). mP<0.05 using Student’s t test.
From a SAR perspective, it is interesting to note the loss of
intrinsic activity when exchanging the piperazine ring of 9a to
a piperidine ring (9b). This effect may be related to a more
preferred conformation of 9a compared to 9b; piperazines are
known to be coplanar between the piperazine ring and the
aromatic ring as the most stable conformation due to sp2
hybridization on the anilinic nitrogen. However, for the
piperidines the most stable conformation is when the piper-
idine ring and the aromatic ring are perpendicular to each
other.58,59 Other differences that may influence the intrinsic
activity are the lower pKa for the piperazines as well as the
character of the π-system, since the anilinic nitrogen will be
partly delocalized. It is also interesting to note that the
3-hydroxy isomers were generally found to bind with higher
in contrast to classic antagonists, increased spontaneous
LMA in partly habituated rats. However, 8 still reduced
LMA in amphetamine-pretreated animals at doses equiva-
lent to those that increased spontaneous activity in partly
habituated rats (Table 3).
The partial agonist 9a had no influence on DOPAC levels
but induced a strong reduction in LMA in normal rats. With
the exception of 10b, which was not tested, compounds
9b-12b all generated a dose-dependent increase in DOPAC
levels (250-300% of control) while inducing moderate
decreases (10a, 11a, 11b, 12a), minor decreases (9b) or, in
an extreme case, increases (12b) in LMA in partly habituated
rats. The increase in LMA displayed by 12b was similar to
that of 8. Compounds 9a-12b where all effective in decreas-
ing amphetamine-induced hyperlocomotion (Table 3), with
the exception of 10b (not tested).
High
Low
and D2
affinity to both D2
than the corresponding
3-methylsulfone isomers. Furthermore, when 9a was com-
pared with 11a, the intrinsic activity was lost, suggesting that
even though a sulfone group can participate in hydrogen
bonding with the hydroxy groups on the dopamine D2
receptor (i.e., serine groups on TM533), this interaction seems
to be less optimal than for the corresponding hydroxy group.
Regarding the N-alkyl group, the lack of intrinsic activity of
the propyl (10a) compared with the benzyl (9a) indicates that
in this series the additional aromatic ring system stabilizes the
high affinity state needed for a functional response.
Discussion
As reported by other authors, the affinity ratio between
High
Low
and D2
D2
correlates positively with the degree of
intrinsic activity. In the current study, agonists, such as
dopamine and apomorphine, bound with strong preference
to the agonist site (D2High) and displayed full intrinsic activity
in the functional assay; in contrast, partial agonists exhibited
intermediateD2High/D2Low ratios andhadlower efficacy inthe
functional assay. The classic dopamine D2 antagonist, halo-
peridol, bound with high affinity to both sites and displayed
no intrinsic activity. However, olanzapine was found to
differentiate between the two sites, showing a 5.2-fold higher
affinity for D2High over D2low. This finding is in line with the
modeling results of Goddard et al., which show that members
of the dibenzodiazepine class (exemplified by clozapine) bind
preferentially to the dopamine agonist pocket, although they
are unable to engender the conformational changes seen with
agonists because of the proposed weaker binding to serine
residues on TM5.38
Lipophilicity often contributes to binding affinity, and in
order to investigate this relation, we plotted clog P values for
Low
the entire set of compounds in Table 2 against the Ki for D2
and D2High (excluding dopamine). As shown in Figure 4, there
Low
was a clear relationship between clog P and affinity for D2
,
indicating that lipophilicity is a factor that contributes to
affinity for dopamine D2 receptors in the inactivated state and
suggesting that such affinity is driven by hydrophobic inter-
actions with hydrophobic residues in the dopamine D2 recep-
High
tor. A similar approach for D2
gave a poor correlation
(R2 = 0.43), which may indicate that more specific interac-
tions, rather than hydrophobic interactions, may be relevant
for agonists and partial agonists. It is also interesting to
note that in the range of lipophilicity and affinity values

2516 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Pettersson et al.
and release of dopamine and its metabolites into the synapse,
whereas the increase in behavioral activity has been regarded
as a consequence of the stimulation of postsynaptic D2
receptors. The opposite effect, with a dose-dependent increase
in dopamine metabolites, reaching a maximum level of ap-
proximately 350% of control levels (Figure 3), and a strong
reduction in LMA (Table 3), was seen with haloperidol,
whereas olanzapine induced a moderate reduction in LMA.
The efficacy on dopamine metabolites was lower with the
partial agonists aripiprazole, (-)-3PPP, and bifeprunox, en-
gendering maximal DOPAC levels of 150-175% of control
levels(Figure3). Thus, levels of DOPAC correlatednegatively
with the level of intrinsic activity (Tables 2 and 3) for agonists,
partial agonists, and antagonists. The partial agonists, much
like the antagonists haloperidol and olanzapine, also induced
a reduction in LMA (Table 3). Compound 9a, which had a
relatively high intrinsic activityin the functional in vitro assay,
was found to be neutral on DOPAC levels but potently
reduced spontaneous LMA. This result indicates that 9a
has a higher intrinsic activity in vivo than the partial agonists
(-)-3-PPP, bifeprunox, and aripiprazole, as determined by
the functional in vitro assay.
Figure 4. Relationship between binding affinities (apparent Ki
of selected compounds to recombinant HEK-293 cells using
[3H]spiperone as a ligand) (pKi) and calculated log P values (clog P).
In contrast to the partial agonist 9a, compounds 9b-12b
lacked intrinsic activity and induced a dose-dependent in-
crease in DOPAC levels similar to that seen with the full D2
antagonists haloperidoland olanzapine(Table3). Despite this
consistent dopamine D2 antagonist neurochemical profile,
compounds 9b-12b displayed a range of effects on sponta-
neous LMA in rats with low psychomotor activity. Some
compounds, such as 11a and 12a, potently inhibited LMA, in
line with what would be predicted from the neurochemical
effects of these compounds (i.e., a maximal increase in DO-
PAC levels). However, compound 12b induced an increase in
LMA but still behaved as a dopamine D2 antagonist in that it
increased DOPAC levels with full efficacy. These effects were
similar to those of the dopaminergic stabilizer, 8, and are
consistent with results reported by others.45,47,64 Independent
of the effects on spontaneous LMA (ranging from stimulation
to reduction), all compounds were effective in blocking
amphetamine-induced hyperactivity (Table 3).
demonstrated for compounds 9b-12b, the attributes of sur-
mountable antagonism and fast off kinetics are pervasive.
Although such a profile might be anticipated to be a product
of low affinity, compounds 9a and 9b both have high affinity
and demonstrate surmountable antagonism and fast off
kinetics. Further, we have previously demonstrated that the
compound N-{[(2S)-5-chloro-7-(methylsulfonyl)-2,3-dihydro-
1,4-benzodioxin-2-yl]methyl}ethanamine (NS30678), which
is equipotent with haloperidol in vitro, also displays this
profile.50 Thus, the surmountable antagonism property is
most probably related to the mode of interaction rather than
a mere low-affinity phenomenon.
Another trend in the data set (i.e., 9a-12b) is that the
piperazines bind with higher affinity to the D2 receptors than
the corresponding piperidines, with the exception of com-
pounds 11a and 11b (Table 2). Even thoughthe piperidines are
more lipophilic than the piperazines when comparing clogP
(3.21 for 9a, 3.91 for 9b, 2.55 for 10a, 3.18 for 10b), we
speculated whether this is relevant under physiological con-
ditions. HPLC is a useful method for comparing the lipophi-
licity of a range of compounds.60,61 We therefore decided to
compare the compounds in a reverse-phase HPLC system
using a water mobile phase at pH 7.4. From the retention time
we then calculated LClogD values for each compound, and
the results clearly show that the piperazines are more lipophilic
than the piperidines (shown as higher LClogD values for the
piperazines; Table 2), which may explain their generally higher
binding affinities. The higher LClogD lipophilicity of the aryl-
piperazines compared with the piperidines is probably explained
by the differences in pKa of the basic nitrogen. Arylpiperazines
display a pKa of ∼6.0-7.5 compared with ∼8.0-9.5 for the
piperidines,62,63 leading to a higher level of ionization of the
arylpiperidines than of the arylpiperazines at physiological pH.
By using in vivo data, we can easily discriminate between
agonists, partial agonists, and antagonists. The typical in vivo
profile demonstrated by a full D2 agonist is a dose-dependent
reduction in tissue levels of dopamine metabolites (DOPAC
and HVA), as shown in Figure 3, plus a pronounced increase
in LMA at high doses (Table 3). The decrease in dopamine
metabolites has been regarded as a consequence of the stimu-
lation of the presynaptic D2 receptors controllingthe synthesis
However, it is not fully explained how compounds that all
behave as antagonists at D2 receptors, based on their effect on
dopamine neurochemistry (i.e., an increase in DOPAC levels),
can induce different effects on behavioral activity in partly
habituated rats. For clarification, the dose at which we
measured behavioral activity is the dose at which a compound
reaches its maximal effect on DOPAC levels. From the data in
Tables 2 and 3 we found, not surprisingly but very interest-
Low
ingly, a clear correlation between the affinity for D2
receptors and the effects on LMA in vivo (Figure 5; R2 =
0.62 for D2Low). A similar correlation was also seen for the
High
D2
receptors (R2 = 0.59; the agonist apomorphine was
excluded). These results indicate that the inhibitory effect on
spontaneous LMA is driven mainly by affinity for the dopa-
High
mine D2 receptors (either D2
or D2Low), independent of
whether the compound is a partial D2 agonist or a compound
with D2 antagonist properties, such as 9b-12b, haloperidol, or
olanzapine. Behavioral activity is known to be influenced by
effects on postsynaptic dopamine D2 receptors, and therefore,
the reduction in LMA is thought to be a consequence of the
displacement of dopamine and the blockade of postsynaptic D2
receptors. This also suggests that the partial agonists aripipra-
zole, bifeprunox, and compound 9a exert their potent reduction
of motor activity by the blockade of postsynaptic D2 receptors.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2517
LMA are diminished. From an SAR perspective, it is inter-
esting to note that the more lipophilic piperazines, 9a, 11a and
12a, induced a stronger reduction in spontaneous LMA in the
partly habituated animals than the corresponding piperidines,
9b, 11b, and 12b. The possible difference in pharmacokinetic
and pharmacodynamic properties can be a concern when
plotting in vivo against in vitro data. However, all compounds
are given subcutaneously to avoid first-pass metabolism and
the most polar compound in this series, 12b, has a brain/
plasma ratio of 3 (unpublished results) and a pronounced
effect on dopamine release and turnover. This supports
the assumption that penetration of the blood-brain barrier
is not an issue within this series. Furthermore, the effects
of pharmacokinetic differences between the compounds are
minimized by studying LMA effects at doses where maximal
DOPAC response is achieved.
One of the hallmarks for dopaminergic stabilizers, such as
12b is that it can increase behavioral activity in animals with
low baseline activity (partly habituated animals) despite hav-
ing full D2 antagonist effects on dopamine synthesis and
metabolism and inhibitory effects on psychostimulant in-
duced behavioral hyperactivity. Differences in extrasynaptic
versus synaptic dopamine neurotransmission,65 partial dopa-
mine receptor agonism,48,66 and the interaction with both
an allosteric and orthosteric site on the dopamine D2 receptor,
resulting in an increased response to dopamine and an
antagonizing dopamine action,49,64 have all been proposed
as mechanisms by which these compounds exert their effects.
Here, we have demonstrated that dopaminergic stabilizers
display unique effects in vitro, such as surmountable anta-
gonism and fast-off kinetics, without intrinsic activity, which
differ from those of classical D2 antagonists and partial
agonists.56 We have also shown that a low affinity for D2
receptors is a prerequisite for an increase in behavioral activity
(Figure 5) among the compounds investigated. As 8 and 12b
both induced a dose-dependent increase in dopamine and
dopamine metabolites (DOPAC) in the striatum, cortex, and
limbic areas, this indicates that these molecules must interact
with D2 autoreceptors, competing with dopamine and leading
to an increase in the synthesis and metabolism of dopamine.
The low affinity and rapid dissociation allow for some fluc-
tuation in the degree of activation of postsynaptic D2 recep-
tors in response to endogenous dopamine surges and, thereby,
do not completely attenuate physiological neurotransmission.
We propose that as affinity increases, the postsynaptic D2
blockade becomes more pronounced and a clear reduction in
behavioral activity is seen (Figure 5). 8 and 12b appear to
exhibit surmountable D2 antagonism and low affinity and act
as state-dependent D2 antagonists, allowing for either an
increase in spontaneous LMA or a decrease in hyperactivity
states, such as that engendered by D-amphetamine.
Figure 5. Relationship between binding affinities (apparent Ki of
selected compounds to recombinant HEK-293 cells using [3H]spi-
perone as a ligand) (pKi) and the logarithm of locomotor activity
expressed as a percentage of control levels (LMA).
Figure 6. Relationship between LClogD (as previously described)
and the logarithm of locomotor activity expressed as a percentage of
control levels (LMA).
Conclusions
It must be emphasized that effects on other receptor systems
could contribute to the influence on motor activity.
We have shown that starting from a novel dopamine D2
partial agonist such as 9a, the modification of physicochemical
properties important for intrinsic activity can lead to the
development of pure dopamine D2 antagonists. By reduction
of the lipophilicity and thereby reduction of the affinity for D2
receptors, compounds with state-dependent effects on LMA
are obtainable. This has led to the discovery of the dopaminer-
gic stabilizer, 12b. It has been shown that this compound
displays surmountable D2 antagonism with fast dissociation,
and on the basis of the chemical motif, we suggest that the mode
of interaction is similar to that of dopamine D2 agonists but
In addition to in vitro affinity, it is also interesting to note
that there was a strong correlation between spontaneous
motor activity and the lipophilicity of the compounds
(R2 = 0.69; Figure 6). This again indicates that potency and
efficacy for D2 antagonism, and thereby reduction in spon-
taneous LMA, are mainly driven by hydrophobic interactions
stabilizing the D2Low state. However, as displayed by 12b, it
seems that by reduction of lipophilicity, it is possible to retain
the effects on dopaminergic neurochemistry (i.e., an increase
in DOPAC levels), while the inhibitory effects on spontaneous

2518 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Pettersson et al.
1
that it lacks essential pharmacophore elements needed for
intrinsic activity and, in addition, lacks the ability to stabilize
the inactive state of the D2 receptor. In conclusion, 12b has a
profile of low lipophilicity, low affinity, surmountable anta-
gonism, and fast receptor dissociation. This permits it to
compete with dopamine in such a way that state-dependent
antagonism of D2 receptor function is attainable while still
allowing attenuated physiological neurotransmission to persist.
This action profile, reflected in vivo by, for example,
efficacy in models of behavioral hyperactivity combined with
a lack of inhibitory motor effects in the normal state, may be
therapeutically useful in the treatment of CNS disorder
related to disturbed activity in central dopaminergic path-
ways, such as movement disorders and schizophrenia.
70 eV) 205 (Mþ, 73), 158 (44), 129 (95), 128 (80), 82 (bp). H
NMR (300 MHz, CDCl3) δ 2.54 (s, 3H), 2.55-2.60 (m, 2H),
3.18-3.24 (m, 2H), 3.51-3.58 (m, 1H), 3.64 (m, 2H), 6.14-6.18
(m, 1H), 7.17-7.24 (m, 2H), 7.27-7.34 (m, 2H).
1-Methoxycarbonyl-4-[3-(methylthio)phenyl]-1,2,3,6-tetrahy-
dropyridine (15). To a solution of 14 (6.1 g, 29.7 mmol) and
triethylamine (5.39 mL, 35.7 mmol) in CH2Cl2 (200 mL) at 0 °C
was added methyl chloroformate (2.53 mL, 32.7 mmol) in
CH2Cl2 (50 mL), and the reaction mixture was stirred for 2 h.
Aqueous Na2CO3 (10%, 100 mL) was added, and the phases
were separated. The aqueous phase was extracted with CH2Cl2
(3 ꢀ 50 mL), and the combined organic phase was washed with
brine, dried (MgSO4), and concentrated in vacuo to give 8.1 g of
crude product. The residue was purified by flash chromatogra-
phy using isooctane/ethyl acetate [1:1 (v/v)] as eluent, afford-
ing pure 15 (7.4 g, 95%). MS m/z (relative intensity, 70 eV)
263 (Mþ, 45), 248 (89), 129 (83), 128 (bp), 59 (96). H NMR
(300 MHz, CDCl3) δ ppm 2.49 (s, 3 H), 2.73 (s, 1 H), 3.64-3.71
(m, 2 H), 3.74 (s, 3 H), 3.77-3.84 (m, 1 H), 4.11 (s, 2 H), 6.05
(s, 1 H), 7.14 (t, J = 7.7 Hz, 2 H), 7.23-7.29 (m, 2 H).
1
Experimental Section
Chemistry General. 1H and 13C NMR spectra were recorded
in MeOH-d6 or CDCl3 at 300 and 75 MHz, respectively, using a
Varian XL 300 spectrometer, or at 400 and 100 MHz, respec-
tively, using a Mercury Plus 400 spectrometer. Chemical shifts
are reported as δ values (ppm) relative to an internal standard
(tetramethylsilane). Low-resolution mass spectra were recorded
on a HP 5970A instrument operating at an ionization potential
of 70 eV. The mass detector was interfaced with a HP5700 gas
chromatograph equipped with a fused silica column (11 m,
0.22 mm i.d.) coated with cross-linked SE-54 (film thickness
0.3 mm, He gas, flow 40 cm/s). Elemental analyses were per-
formed by MikroKemi AB, Uppsala, Sweden. Melting points
were determined with a point microscope (Reichert Thermovar)
and are uncorrected. For flash chromatography, silica gel 60
(0.040-0.063 mm, VWR, no. 109385) was used. The amine
products were converted to the corresponding salts by dissol-
ving the free base in EtOH and adding 1 equiv of the acid
(fumaric or oxalic) or ethanolic HCl solution. The solvent was
removed and azeotroped with absolute EtOH in vacuo followed
by recrystallization from appropriate solvents. Purity of all
target compounds where assessed as greater than 95% by
elemental analysis (C, H, N).
1-Methoxycarbonyl-4-[3-(methylsulfonyl)phenyl]-1,2,3,6-tetra-
hydropyridine (16). A solution of 15 (6.28 g, 23.87 mmol)
in CH2Cl2 (300 mL) was cooled to 0 °C, and m-CPBA (12.2 g,
52.5 mmol) was added in portions over a period of 30 min. The
mixture was stirred at 0 °C for 1.5 h and then at ambient
temperature for 1 h. Aqueous Na2CO3 (10%, 100 mL) was
added, and the phases were separated. The aqueous phase was
extracted with CH2Cl2 (3 ꢀ 50 mL) and the combined organic
phase was washed with brine, dried (MgSO4), and concentra-
ted in vacuo to give 8.6 g of crude product that was used in
the subsequent step without any further purification. MS m/z
(relative intensity, 70 eV) 295 (Mþ, 19), 280 (56), 129 (70), 128
(89), 59 (bp).
1-Methoxycarbonyl-4-[3-(methylsulfonyl)phenyl]piperidine (17).
To a solution of 16 (8.06 g, 27.3 mmol) in MeOH (160 mL) was
added concentrated HCl (8 mL) and Pd/C (1.5 g) under N2, and
the reaction mixture was hydrogenated under H2 (50 psi) for 15 h.
Filtration through Celite and evaporation of the filtrate afforded
8.6 g of crude product as the HCl salt. Purification with flash
chromatography using CH2Cl2/MeOH [30:1 (v/v)] as eluent
afforded pure 17 (6.73 g, 83%). MS m/z (relative intensity,
Detailed Synthetic Procedures. 1-tert-Butoxycarbonyl-4-hy-
droxy-4-[3-(methylthio)phenyl]piperidine (13). To a solution of
3-bromothioanisole (4.25 g, 20.9 mmol) in dry THF (30 mL) at
-78 °C was added n-buthyllithium in hexane (2.5 M, 9.3 mL,
23.2 mmol). The mixture was stirred at -78 °C under an N2
atmosphere for 30 min and then allowed to warm to -20 °C for
2 min. After the mixture was again cooled to -78 °C, 1-Boc-
4-piperidone (4.38 g, 22.0 mmol) in dry THF (20 mL) was added
via syringe. The solution was allowed to warm to 20 °C and
stirred for an additional 10 min. The reaction mixture was then
diluted with aqueous NH4Cl, and the phases were separated.
The aqueous phase was extracted with ethyl acetate (3 ꢀ 50 mL)
and the combined organic phase was washed with brine, dried
(MgSO4), and concentrated in vacuo to give 6.8 g of crude
product. The residue was purified by flash chromatography
using CH2Cl2/MeOH [19:1 (v/v)] as eluent, affording 13 as a
gum (4.9 g, 72%). MS m/z (relative intensity, 70 eV) 323 (Mþ,
1
70 eV) 297 (Mþ, 54), 282 (62), 238 (bp), 115 (92), 56 (93). H
NMR (300 MHz, CDCl3) δ 1.58-1.77 (m, 2 H), 1.79-1.89 (m,
2 H), 2.62-2.81 (m, 3 H), 3.07-3.16 (m, 5 H), 7.52-7.61 (m, 2 H),
7.76-7.87 (m, 2 H).
4-[3-(Methylsulfonyl)phenyl]piperidine (18). To a solution
of 17 (5.1 g, 17.2 mmol) in absolute EtOH (40 mL), HCl (8 M,
100 mL) was added. The mixture was refluxed for 24 h. The
EtOH was evaporated, and the aqueous phase was made basic
with NaOH (5 M). The aqueous phase was extracted with ethyl
acetate (3 ꢀ 50 mL) and the combined organic phase was washed
with brine, dried (MgSO4), concentrated in vacuo, and purified
by flash chromatography using ethyl acetate/MeOH [1:1 (v/v)]
as eluent to give 3.5 g (86%) of 18. MS m/z (relative intensity,
70 eV) 239 (Mþ, 59), 238 (50), 69 (20), 57 (79), 56 (bp). 1H NMR
(300 MHz, CDCl3) δ ppm 1.66 (dd, J = 12.6, 4.20 Hz, 2 H), 1.85
(s, 2 H), 2.78-2.93 (m, 3 H), 3.06 (s, 3 H), 3.73 (s, 3 H),
7.44-7.57 (m, 2 H), 7.71-7.85 (m, 2 H).
1
10), 267 (17), 178 (12), 57 (bp), 56 (15). H NMR (300 MHz,
CDCl3) δ 1.83 (s, 9 H), 2.05 (d, J = 12.4 Hz, 4 H), 2.23-2.40 (m,
2 H), 2.84 (s, 3H), 3.57 (t, J = 12.0 Hz, 2H), 4.37 (br s, 1H), 7.50
(d, J = 7.3 Hz, 1H), 7.55 (d, J = 1.5 Hz, 2 H), 7.62 (d, J=7.6 Hz,
1H), 7.74 (d, J=1.5 Hz, 1H).
4-[3-(Methylthio)phenyl]-1,2,3,6-tetrahydropyridine (14). To a
solution of 13 (10.1 g, 31.2 mmol) in CH2Cl2 (250 mL) was added
TFA (70 mL), and the mixture was refluxed for 48 h. After
cooling, the reaction mixture was poured onto NaOH (5 M,
150 mL) and the phases were separated. The aqueous phase was
extracted with CH2Cl2 (3 ꢀ 50 mL), and the combined organic
phase was washed with brine, dried (MgSO4), and concentrated
in vacuo to give 5.77 g (90%) of 14. MS m/z (relative intensity,
4-[3-(Methylsulfonyl)phenyl]-1-propylpiperidine (12b). Com-
pound 18 (3.5 g, 14.7 mmol) was dissolved in acetonitrile (200
mL), and iodopropane (1.72 mL, 17.7 mmol) and K2CO3 (5.0 g,
36.9 mmol) were added. The reaction mixture was allowed to
reflux for 15 h. after the mixture was cooled to ambient
temperature, aqueous Na2CO3 (10%, 100 mL) was added, and
the phases were separated. The aqueous phase was extracted
with ethyl acetate (3 ꢀ 50 mL) and the combined organic phase
was washed with brine, dried (MgSO4), and concentrated in
vacuo to give 4.2 g of crude product. Purification with flash
chromatography using CH2Cl2/MeOH [1:1 (v/v)] as eluent afforded
pure 12b (3.28 g, 79%). MS m/z (relative intensity, 70 eV) 281

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6 2519
(Mþ, 5), 252 (bp), 129 (20), 115 (20), 70 (25). 1H NMR (300 MHz,
CDCl3) δ ppm 0.96 (t, J = 7.3 Hz, 3 H), 1.53-1.64 (m, 2 H), 1.89
(dd, J= 9.6, 3.54 Hz, 4 H), 2.03-2.14(m, 2H), 2.31-2.41(m, 2H),
2.64(ddd,J= 15.4, 5.7, 5.5 Hz, 1 H), 3.06-3.15(m,5H),7.51-7.58
(m, 2 H), 7.78-7.86 (m, 2 H). 13C NMR (75 MHz, CDCl3) δ ppm
11.98, 20.18, 33.29, 42.59, 44.43, 54.06, 60.93, 124.99, 125.74, 129.39,
132.04, 148.28. The amine was converted to the HCl salt and
recrystallized in EtOH/diethyl ether: mp 212-214 °C. Anal.
(C15H24ClNO2S) C, H, N.
Spangler, T.; Brennan, J. A.; Piesla, M.; Mazandarani, H.; Cockett,
M. I.; Ochalski, R.; Coupet, J.; Andree, T. H. New generation
dopaminergic agents. 1. Discovery of a novel scaffold which
embraces the D2 agonist pharmacophore. Structure-activity re-
lationships of a series of 2-(aminomethyl)chromans. J. Med. Chem.
1997, 40, 4235–4256.
(12) Clark, D.; Hjort, S.; Carlsson, A. Dopamine receptor agonists:
mechanisms underlying autoreceptor selectivity II. Theoretical
considerations. J. Neural Transm. 1985, 62, 171–207.
(13) Casey, D. E.; Keepers, G. A. Neuroleptic side effects: acute
extrapyramidal syndromes and tardive dyskinesia. Psychopharma-
col. Ser. 1988, 5, 74–93.
(14) Lahti, A. C.; Weiler, M. A.; Corey, P. K.; Lahti, R. A.; Carlsson,
A.; Tamminga, C. A. Antipsychotic properties of the partial
dopamine agonist (-)-3-(3-hydroxyphenyl)-N-n-propylpiperidine
(preclamol) in schizophrenia. Biol. Psychiatry 1998, 43, 2–11.
(15) Petrie, J. L.; Saha, A. R.; McEvoy, J. P. Aripiprazole, a new
atypical antipsychotic: phase II clinical trial results. Eur. Neuro-
psychopharmacol. 1997, 7, 227.
(16) Carlsson, A.; Tamminga, C. A. Partial dopamine agonists and
dopaminergic stabilizers, in the treatment of psychosis. Curr. Drug
Targets: CNS Neurol. Disord. 2002, 1, 141–147.
Acknowledgment. We thank Maria Gullme, Anna Sandahl,
˚
Mikael Andersson, and Hakan Schyllander for help with
the synthetic chemistry, and Elisabeth Ljung, Marianne
€
Thorngren, Kirsten Sonniksen, Theresa Andreasson, Boel
Svanberg, and Therese Carlsson for their work with beha-
ꢀ
vioral and neurochemical experiments and analyses. We also
thank Tino Dyhring and Elsebet Østergaard Nielsen for
providing the in vitro functional data (fluorescent imaging
plate reader), and Anna-Carin Jansson, Lars Swanson, and
Peder Svensson for their help in producing LClogD values.
For performance of calculations of Ki values and statistics, we
thank Andreas Stansvik and Ylva Sunesson. For review of
the manuscript, we thank Jonas Karlsson, Peder Svensson,
Naheed Mirza, Kristina Luthman, and Palle Christophersen.
(17) Broadley, K. J.; Kelly, D. R. Muscarinic receptor agonists and
antagonists. Molecules 2001, 6, 142–193.
(18) Esbenshade, T. A.; Fox, G. B.; Cowart, M. D. Histamine H3
receptor antagonists: preclinical promise for treating obesity and
cognitive disorders. Mol. Interventions 2006, 6, 77–88.
(19) Gualtieri, F. Cholinergic receptors and neurodegenerative
diseases. Pharm. Acta Helv. 2000, 74, 85–89.
(20) Kubinyi, H. Chemical similarity and biological activites. J. Braz.
Chem. Soc. 2002, 13, 717–726.
(21) Smith, J. H. Introduction to the Principles of Drug Design and
Action, 3rd ed.; Harwood Academic Publishers: London, 1998; pp
387-433.
Note Added after ASAP Publication. This paper was
published ASAP on February 15, 2010 with errors in Figures
5 and 6. The revised version was published on February 23,
2010.
(22) Parker, M. A.; Kurrasch, D. M.; Nichols, D. E. The role of lipo-
philicity in determining binding affinity and functional activity for
5-HT2A receptor ligands. Bioorg. Med. Chem. 2008, 16, 4661–4669.
(23) Runyon, S. P.; Mosier, P. D.; Roth, B. L.; Glennon, R. A.;
Westkaemper, R. B. Potential modes of interaction of 9-amino-
methyl-9,10-dihydroanthracene (AMDA) derivatives with the
5-HT2A receptor: a ligand structure-affinity relationship, recep-
tor mutagenesis and receptor modeling investigation. J. Med.
Chem. 2008, 51, 6808–6828.
Supporting Information Available: Experimental details of the
synthesis of 9a-12a; biological methods and methods of pre-
paring LClogD values. This material is available free of charge
via the Internet at http://pubs.acs.org.
(24) Ariens, E. J. Drug Design; Academic Press: New York, 1971; Vol. 1,
pp 162-193.
References
(25) Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis
in Chemistry and Biology; Wiley: New York, 1979.
(1) Johnson, G. L.; Dhanasekaran, N. The G-protein family and their
interaction with receptors. Endocr. Rev. 1989, 10, 317–331.
(2) Seeman, P. Atypical antipsychotics: mechanism of action. Can. J.
Psychiatry 2002, 47, 27–38.
(3) Seeman, P. An Update of Fast-Off Dopamine D2 Atypical Anti-
psychotics. Am. J. Psychiatry 2005, 162, 1984–1985.
(4) George, S. R.; Watanabe, M.; Di Paolo, T.; Falardeau, P; Labrie,
F.; Seeman, P. The functional state of the dopamine receptor in the
anterior pituitary is in the high affinity form. Endocrinology 1985,
117, 690–697.
(5) Sibley, D. R.; De Lean, A.; Creese, I. Anterior pituitary dopamine
receptors. Demonstration of interconvertible high and low affinity
states of the D-2 dopamine receptor. J. Biol. Chem. 1982, 257,
6351–6361.
(26) Cho, W.; Taylor, L. P.; Mansour, A.; Akil, H. Hydrophobic
residues of the D2 dopamine receptor are important for binding
and signal transduction. J. Neurochem. 1995, 65, 2105–2215.
(27) Cox, B. A.; Henningsen, R. A.; Spanoyannis, A.; Neve, R. L.;
Neve, K. A. Contributions of conserved serine residues to the
interactions of ligands with dopamine D2 receptors. J. Neurochem.
1992, 59, 627–635.
(28) Mansour, A.; Meng, F.; Meador-Woodruff, J. H.; Taylor, L. P.;
Civelli, O.; Akil, H. Site-directed mutagenesis of the human
dopamine D2 receptor. Eur. J. Pharmacol. 1992, 227, 205–214.
(29) Shi, L.; Javitch, J. A. The binding site of aminergic G protein-
coupled receptors: the transmembrane segments and second extra-
cellular loop. Annu. Rev. Pharmacol. Toxicol. 2002, 42, 437–467.
(30) Teeter, M. M.; Froimowitz, M.; Stec, B.; DuRand, C. J. Homology
modeling of the dopamine D2 receptor and its testing by docking of
agonists and tricyclic antagonists. J. Med. Chem. 1994, 37, 2874–2888.
(31) Wiens, B. L.; Nelson, C. S.; Neve, K. A. Contribution of serine
residues to constitutive and agonist-induced signaling via the D2S
dopamine receptor: evidence for multiple, agonist-specific active
conformations. Mol. Pharmacol. 1998, 54, 435–444.
(32) Wilcox, R. E.; Huang, W.-H.; Brusniak, M.-Y. K.; Wilcox, D. M.;
Pearlman, R. S.; Teeter, M. M.; DuRand, C. J.; Wiens, B. L.; Neve,
K. A. CoMFA-based prediction of agonist affinities at recombi-
nant wild type versus serine to alanine point mutated D2 dopamine
receptors. J. Med. Chem. 2000, 43, 3005–3019.
(33) Yashar, M.; Kalani, S.; Vaidehi, N.; Hall, S. E.; Trabanino, R. J.;
Freddolino, P. L.; Kalani, M. A.; Floriano, W. B.; Tak Kam,
V. W.; Goddard, W. A., III. The predicted 3D structure of the
human D2 dopamine receptor and the binding site and binding
affinities for agonists and antagonists. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 3815–3820.
(6) Payne, S. L.; Johansson, A. M.; Strange, P. G. Mechanisms of
ligand binding and efficacy at the human D2(short) dopamine
receptor. J. Neurochem. 2002, 82, 1106–1117.
(7) Seeman, P.; Tallerico, T.; Ko, F. Dopamine displaces [3H]domperi-
done from high-affinity sites of the dopamine D2 receptor, but
not [3H]raclopride or [3H]spiperone in isotonic medium: implica-
tions for human positron emission tomography. Synapse 2003, 49,
209–215.
(8) Kenakin, T. Collateral efficacy in drug discovery: taking advantage
of the good (allosteric) nature of 7TM receptors. Trends Pharma-
col. Sci. 2007, 28, 407–415.
(9) Lahti, R. A.; Figur, L. M.; Piercey, M. F.; Ruppel, P. L.; Evans,
D. L. Intrinsic activity determinations at the dopamine D2 guanine
nucleotide-binding protein-coupled receptor: utilization of recep-
tor state binding affinities. Mol. Pharmacol. 1992, 42, 432–439.
˚
(10) Malmberg, A.; Mohella, N.; Backlund Hook, B.; Johansson,
€€
A. M.; Hacksell, U.; Nordvall, G. Interactions of ligands with
active and inactive conformations of the dopamine D2 receptor.
Eur. J. Pharmacol. 1998, 346, 299–307.
(34) Vilardaga, J.-P. Switching modes for G protein-coupled receptor
activation. Nat. Chem. 2006, 2, 395–396.
(35) Ballesteros, J. A.; Jensen, A. D.; Liapakis, G.; Rasmussen, S.; Shi,
L.; Gether, U.; Javitch, J. A. Activation of the 2-adrenergic
(11) Mewshaw, R. E.; Kavanagh, J.; Stack, G.; Marquis, K. L.; Shi, X.;
Kagan, M. Z.; Webb, M. B.; Katz, A. H.; Park, A.; Kang, Y. H.;
Abou-Gharbia, M.; Scerni, R.; Wasik, T.; Cortes-Burgos, L.;

2520 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 6
Pettersson et al.
receptor involves disruption of an ionic lock between the cyto-
plasmic ends of transmembrane segments 3 and 6. J. Biol. Chem.
2001, 276, 29171–29177.
stabilizers pridopidine (ACR16) and (-)-OSU6162 display dop-
amine D2 receptor antagonism and fast receptor dissociation pro-
perties. Eur. J. Pharmacol. 2010, 628 (1-3), 19–26.
(51) McGaughey, G. B.; Mewshaw, R. E. Application of comparative
molecular field analysis to dopamine D2 partial agonists. Bioorg.
Med. Chem. Lett. 1999, 7, 2453–2456.
(36) Hubbell, W. L.; Altenbach, C.; Hubbell, C. M.; Khorana, H. G.
Rhodopsin structure, dynamics, and activation: a perspective
from crystallography, site-directed spin labeling, sulfhydryl
reactivity, and disulfide cross-linking. Adv. Protein Chem.
2003, 63, 243–290.
(37) Yao, X.; Parnot, C.; Deupi, X.; Ratnala, V. R. P.; Swaminath, G.;
Farrens, D.; Kobilka, B. Coupling ligand structure to specific
conformational switches in the beta(2)-adrenoceptor. Nat. Chem.
Biol. 2006, 2, 417–422.
(52) Mewshaw, R. E.; Husbands, M.; Gildersleeve, E. S.; Webb, M. B.;
Shi, X.; Mazandarani, H.; Cocke, M. I.; Ochalski, R.; Brennan,
J. A.; Abou-Gharbia, M.; Marquis, K.; McGaughey, G. B.;
Coupet, J.; Andree, T. H. New generation dopaminergic agents. 2.
Discovery of 3-OH-phenoxyethylamine and 3-OH-N1-phenyl-
piperazine. Bioorg. Med. Chem. Lett. 1998, 8, 295–300.
(53) Bartholomew, C. H.; Agrawal, P. K.; Katzer, J. R. Sulphur
poisoning of metals. Adv. Catal. 1982, 31, 135–242.
(54) Burris, K. D.; Molski, T. F.; Xu, C.; Ryan, E.; Tottori, K.; Kikuchi,
T.; Yocca, F. D.; Molinoff, P. B. Aripiprazole, a novel antipsycho-
tic, is a high-affinity partial agonist at human dopamine D2
receptors. J. Pharmacol. Exp. Ther. 2002, 302, 381–389.
(55) Tadori, Y.; Miwa, T.; Tottori, K.; Burris, K. D.; Stark, A.; Mori,
T.; Kikuchi, T. Aripiprazole’s low intrinsic activities at human
dopamine D2L and D2S receptors render it a unique antipsychotic.
Eur. J. Pharmacol. 2005, 515, 10–19.
(56) Tadori, Y.; Kitagawa, H.; Forbes, R. A.; McQuade, R. D.; Stark,
A.; Kikuchi, T. Differences in agonist/antagonist properties at
human dopamine D2 receptors between aripiprazole, bifeprunox
and SDZ 208-912. Eur. J. Pharmacol. 2007, 574, 103–111.
(57) Hacksell, U.; Arvidsson, L. E.; Svensson, U.; Nilsson, J. L. G.;
Sanchez, D.; Wikstroem, H.; Lindberg, P.; Hjorth, S.; Carlsson, A.
3-Phenylpiperidines. Central dopamine-autoreceptor stimulating
activity. J. Med. Chem. 1981, 24, 1475–1482.
(38) Goddard, W. A., III; Abrol, R. 3-Dimensional structure of G
protein-coupled receptors and binding sites of agonists and
antagonists. J. Nutr. 2007, 137, 1528–1538.
(39) Dowd, C. S.; Herrick-Davis, K.; Egan, C.; DuPre, A.; Smith, C.;
Teitler, M.; Glennon, R. A. 1-[4-(3-Phenylalkyl)phenyl]-2-aminopro-
panes as 5-HT2A partial agonists. J. Med. Chem. 2000, 43, 3074–3084.
(40) Seggel, M. R.; Yousif, M. Y.; Lyon, R. A.; Titeler, M.; Roth, B. L.;
Suba, E. A.; Glennon, R. A. A structure-affinity study of the
binding of 4-substituted analogues of 1-(2,5-dimethoxyphenyl)-2-
aminopropanaet 5-HT2 serotonin receptors. J. Med. Chem. 1990,
33, 1032–1036.
ꢀ
(41) Assie, M. B.; Koek, W. Effects of 5-HT1A receptor antagonists on
hippocampal 5-hydroxytryptamine levels: (S)-WAY100135, but
not WAY100635, has partial agonist properties. Eur. J. Pharmacol.
1996, 304, 15–21.
(42) Mattsson, C.; Sonesson, C.; Sandahl, A.; Greiner, H. E.; Gassen,
M.; Plaschke, J.; Leibrock, J.; Bottcher, H. 2-Alkyl-3-(1,2,3,6-
tetrahydropyridin-4-yl)-1H-indoles as novel 5-HT6 receptor ago-
nists. Bioorg. Med. Chem. Lett. 2005, 15, 4230–4234.
(43) Clark, D.; Hjort, S.; Carlsson, A. Dopamine-receptor agonists:
mechanisms underlying autoreceptor selectivity. J. Neural Transm.
1985, 62, 1–52.
(58) Dijkstra, G. D. H. Conformational analysis of 1-arylpiperazines
and 4-arylpiperidines. Rec. Trav. Chim. Pays-Bas 1993, 112, 151–
160.
(59) McGaughey, G. B.; Mewshaw, R. E. Molecular Modeling and the
Design of Dopamine D2 Partial Agonists. http://www.netsci.org/
Science/Compchem/feature20.html.
(44) Sonesson, C.; Waters, N.; Svensson, K.; Carlsson, A.; Smith, M.
€
W.; Piercey, M. F.; Meier, E.; Wikstrom, H. Substituted 3-phe-
(60) Caffieri, S. Reversed-phase high-performance liquid chromato-
graphy (RP-HPLC) determination of lipophilicity of furocou-
marins: relationship with DNA interaction. J. Pharm. Sci. 2001,
90, 732–739.
nylpiperidines: new centrally acting dopamine autoreceptor an-
tagonists. J. Med. Chem. 1993, 36, 3188–3196.
(45) Sonesson, C.; Lin, C.-H.; Hansson, L.; Waters, N.; Svensson, K.;
€
Carlsson, A.; Smith, M. W.; Wikstrom, H. Substituted (S)-phe-
(61) Leopoldo, M.; Lacivita, E.; De Giorgio, P.; Fracasso, C.; Guzzetti,
S.; Caccia, S.; Contino, M.; Colabufo, N. A.; Berardi, F.; Perrone,
R. Structural modifications of N-(1,2,3,4-tetrahydronaphthalen-1-
yl)-4-aryl-1-piperazinehexanamides: influence on lipophilicity and
5-HT7 receptor activity. Part III. J. Med. Chem. 2008, 51, 5813–
5822.
nylpiperidines and rigid congeners as preferential dopamine
autoreceptor antagonists: synthesis and structure-activity rela-
tionships. J. Med. Chem. 1994, 37, 2735–2753.
(46) Carlsson, M. L.; Nilsson, M.; Carlsson, A. Schizophrenia: from
dopamine to glutamate and back. Curr. Med. Chem. 2004, 11, 267–
277.
(62) ACD/Labs, version 10; Advanced Chemistry Devlopment, Inc.:
Toronto, Ontario, Canada, 2007.
(63) Leopoldo, M.; Lacivita, E.; De Giorgio, P.; Berardi, F.; Perrone, R.
Determination of 1-aryl-4-propylpiperazine pKa values: the sub-
stituent on aryl modulates basicity. Bioorg. Med. Chem. 2009, 17,
1339–1344.
(64) Rung, J. P.; Rung, E.; Helgeson, L.; Johansson, A. M.; Svensson,
K.; Carlsson, A.; Carlsson, M. L. Effects of (-)-OSU6162 and
ACR16 on motor activity in rats, indicating a unique mechanism of
dopaminergic stabilization. J. Neural Transm. 2008, 115, 899–908.
(65) Carlsson, A.; Carlsson, M. L. A dopaminergic deficit hypothesis of
schizophrenia: the path to discovery. Dial. Clin. Neurosci. 2006, 8,
137–142.
(66) Seeman, P.; Tokita, K.; Matsumoto, M.; Matsuo, A.; Sasamata,
M.; Miyata, K. D. The dopaminergic stabilizer ASP2314/ACR16
selectively interacts with D2High receptors. Synapse 2009, 8, 930–934.
(47) Natesan, S.; Svensson, K. A.; Reckless, G. E.; Nobrega, J. N.;
Barlow, K. B. L.; Johansson, A. M.; Kapur, S. The dopa-
mine stabilizers (S)-(-)-(3-methanesulfonyl-phenyl)-1-propyl-pi-
peridine [(-)-OSU6162] and 4-(3-methanesulfonylphenyl)-1-pro-
pyl-piperidine (ACR16) show high in vivo D2 receptor occupancy,
antipsychotic-like efficacy, and low potential for motor side effects
in the rat. J. Pharmacol. Exp. Ther. 2006, 318, 810–818.
(48) Seeman, P.; Guan, H. C. Dopamine partial agonist action of
(-)-OSU6162 is consistent with dopamine hyperactivity in psycho-
sis. Eur. J. Pharmacol. 2006, 557, 151–153.
(49) Lahti, R. A.; Tamminga, C. A.; Carlsson, A. Stimulating and
inhibitory effects of the dopamine “stabilizer” (-)-OSU6162 on
dopamine D2 receptor function in vitro. J. Neural Transm. 2007,
114, 1143–1146.
(50) Dyhring, T.; Nielsen, E. Ø.; Sonesson, C.; Pettersson, F.; Karlsson,
J.; Svensson, P.; Christophersen, P.; Waters, N. The dopaminergic